Knowledge-Guided Gene Panel Selection for Label-Free Single-Cell RNA-Seq Data: A Reinforcement Learning Perspective
- 1 November 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Biology and Bioinformatics
- Vol. 22 (6) , 3041-3054
Abstract
Gene panel selection aims to identify the most informative genomic biomarkers in label-free genomic datasets. Traditional approaches, which rely on domain expertise, embedded machine learning models, or heuristic-based iterative optimization, often introduce biases and inefficiencies, potentially obscuring critical biological signals. To address these challenges, we present an iterative gene panel selection strategy that harnesses ensemble knowledge from existing gene selection algorithms to establish preliminary boundaries or prior knowledge, which guide the initial search space. Subsequently, we incorporate reinforcement learning (RL) through a reward function shaped by expert behavior, enabling dynamic refinement and targeted selection of gene panels. This integration mitigates biases stemming from initial boundaries while capitalizing on RL's stochastic adaptability. Comprehensive comparative experiments, case studies, and downstream analyses demonstrate the effectiveness of our method, highlighting its improved precision and efficiency for label-free biomarker discovery. Our results underscore the potential of this approach to advance single-cell genomics data analysis.Keywords
Funding Information
- Strategic Priority Research Program of the Chinese Academy of Sciences (XDA0460101)
- National Natural Science Foundation of China (92470204)
- Beijing Natural Science Foundation (4254089)
- SciHorizon Platform via the Fundamental Research Project of CNIC (E455230401)
This publication has 62 references indexed in Scilit:
- scCompass: An Integrated Multi‐Species scRNA‐seq Database for AI‐ReadyAdvanced Science, 2025
- Voting from Nearest Tasks: Meta-Vote Pruning of Pre-trained Models for Downstream Tasks2023
- Highly Regional Genes: graph-based gene selection for single-cell RNA-seq dataJournal of Genetics and Genomics, 2022
- Actor-Critic Models and the A3C2019
- Comprehensive Integration of Single-Cell DataCell, 2019
- Mapping the Mouse Cell Atlas by Microwell-SeqCell, 2018
- Spatial reconstruction of single-cell gene expression dataNature Biotechnology, 2015
- Recursive feature elimination with random forest for PTR-MS analysis of agroindustrial productsChemometrics and Intelligent Laboratory Systems, 2006
- Principal components analysis (PCA)Computers & Geosciences, 1993
- Silhouettes: A graphical aid to the interpretation and validation of cluster analysisJournal of Computational and Applied Mathematics, 1987